The proposed research merges theory and intervention research in the behavioral and social sciences to address a major public health problem in the U.S. -- excessive alcohol consumption on university campuses and concomitant risks for alcohol-impaired driving (i.e., driving while intoxicated or DWI). From a theoretical perspective, the two year project will assess the validity of a theory-driven "actively caring model" designed by the PI and his students to predict who will become intervention agents (i.e., individuals who will volunteer to apply a particular intervention technique to reduce the probability of excessive alcohol consumption or DWI). The research will also evaluate the extent that certain verbal report measures of person factors (i.e. specific intentions, risky lifestyles, and certain personality characteristics) predict the amount of alcohol consumption at a fraternity party. From a practical problem-solving perspective, the research will develop and test various intervention techniques to reduce the probability of excessive alcohol consumption at university fraternity parties and subsequent DWI risk; and during follow-up research, a campuswide dissemination of the intervention techniques will be evaluated and training materials will be developed for large-scale application (e,g., on college and university campuses throughout the U.S.). Prior to eight fraternity parties, students' drinking intentions, lifestyles, and person characteristics (i.e., self-esteem, optimism, personal control, group cohesion, sensation seeking) will be measured. Before and after the fraternity parties, students' blood alcohol concentration (BAC) will be objectively assessed with a breathalyzer. During the fraternity parties, students' participation in various intervention techniques will be systematically observed. The impact of a refined intervention process for reducing the risk of DWI will be assessed with both within-subject and between subject comparisons. That is, two fraternities will have two successive parties, one with the intervention process and the other as a control (with a balanced AB vs. BA format). For the campuswide evaluation, the Office of Student Affairs will help us disseminate the intervention techniques to all fraternities (n=34) and sororities (n=16). Through interviews, questionnaires, and focus group meetings we will assess the impact and social validity of the intervention process developed from the proposed project.